AI Interaction Data Capture helps you automatically collect and organise important customer details during conversations, without slowing the conversation down. Instead of rigid scripts, our AI tools review the transcript in real time, pulling out key information like name, account number, date of birth, or even the reason for the conversation.
This means your team always has the right data on hand for reporting, CRM updates, and follow-ups, all captured seamlessly in the background.
This feature currently uses OpenAI to process transcripts for data extraction. Only conversation transcripts are shared, no metadata. If your policies restrict transcript sharing, do not enable this feature.
▶️ How AI Interaction Data Capture Works
When enabled, AI will:
- Read conversation transcripts (voice, video, or messaging) in real time.
- Extract predefined data fields you’ve set up (e.g. Name, Policy Number, Summary).
- Update the interaction data live as the conversation continues.
- Send captured data to your CRM or data warehouse automatically after the interaction.
Agents can always see captured data in the console and double-check or edit it during wrap-up.
🎣 Configuring Your AI Interaction Data Capture

Step 1: Open AI Interaction Data Config
Head to ‘Settings’ in the left-hand menu, then under “Feature Management” you will find AI Interaction Data Capture Configs.
You can also use the search bar to quickly access.
Click “Create AI Interaction Data Capture Config”
Step 2: Complete the Config Form
- Name your configuration so you can identify it later.
- Select the queues this configuration should apply to.
- Add or edit fields for the data you want to capture (e.g. Name, Policy Number, Date of Birth).
- In each field’s description, clearly tell the AI what to look for (e.g. “The customer’s full name. If only the first name is given, capture just that”).
- Repeat this process to add as many fields as you need.
- When finished, click Save.
A tip for writing your descriptions is to imagine you are explaining what to look out for to a brand new employee with very literal thinking.


Step 3: Watch It Work
Once your configuration is live, the AI will begin filling in your chosen fields as conversations happen. Agents can see the captured data in real time under the Information tab and in the Interaction Log.
Don’t be afraid to experiment! You can capture more than just names and emails, try creative fields that support your business goals, like follow-ups required or objections raised.